Recording the activity of a software application during execution typically results in rich, vast amounts of logging information, particularly when the software application is executed in a massively parallel environment over a large cluster of host computers. The logging information can be retained in files across the cluster of host computers and/or in other files within an archival storage system. As such, access to specific portions of the logging information commonly entails searching for information across multiple host computers and/or multiple files within the archival systems. Conventional mechanisms for searching logging information can be tedious to implement and are usually time consuming, which in turn can hinder the troubleshooting, maintenance, and/or development of large, complex software applications. Much remains to be improved in the searching of logging information.